CariGANs: Unpaired Photo-to-Caricature Translation
arXiv:1811.00222 · doi:10.1145/3272127.3275046
Abstract
Facial caricature is an art form of drawing faces in an exaggerated way to convey humor or sarcasm. In this paper, we propose the first Generative Adversarial Network (GAN) for unpaired photo-to-caricature translation, which we call "CariGANs". It explicitly models geometric exaggeration and appearance stylization using two components: CariGeoGAN, which only models the geometry-to-geometry transformation from face photos to caricatures, and CariStyGAN, which transfers the style appearance from caricatures to face photos without any geometry deformation. In this way, a difficult cross-domain translation problem is decoupled into two easier tasks. The perceptual study shows that caricatures generated by our CariGANs are closer to the hand-drawn ones, and at the same time better persevere the identity, compared to state-of-the-art methods. Moreover, our CariGANs allow users to control the shape exaggeration degree and change the color/texture style by tuning the parameters or giving an example caricature.
arXiv commentTo appear at SIGGRAPH Asia 2018
References in corpus (6)
- Adam: A Method for Stochastic Optimization
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Progressive Growing of GANs for Improved Quality, Stability, and Variation
- Toward Multimodal Image-to-Image Translation
- Unsupervised Image-to-Image Translation Networks
- DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
Cited by in corpus (25)
- StyleFlow: Attribute-conditioned Exploration of StyleGAN-Generated Images using Conditional Continuous Normalizing Flows
- TileGAN: Synthesis of Large-Scale Non-Homogeneous Textures
- StyleCariGAN: Caricature Generation via StyleGAN Feature Map Modulation
- 3D-CariGAN: An End-to-End Solution to 3D Caricature Generation from Face Photos
- AgileAvatar: Stylized 3D Avatar Creation via Cascaded Domain Bridging
- MW-GAN: Multi-Warping GAN for Caricature Generation with Multi-Style Geometric Exaggeration
- TransGaGa: Geometry-Aware Unsupervised Image-to-Image Translation
- GANHopper: Multi-Hop GAN for Unsupervised Image-to-Image Translation
- Modeling Caricature Expressions by 3D Blendshape and Dynamic Texture
- Geometric Style Transfer
- AniGAN: Style-Guided Generative Adversarial Networks for Unsupervised Anime Face Generation
- Landmark Detection and 3D Face Reconstruction for Caricature using a Nonlinear Parametric Model
- EDIT: Exemplar-Domain Aware Image-to-Image Translation
- WarpGAN: Automatic Caricature Generation
- RoboCoDraw: Robotic Avatar Drawing with GAN-based Style Transfer and Time-efficient Path Optimization
- GuidedStyle: Attribute Knowledge Guided Style Manipulation for Semantic Face Editing
- 3D Magic Mirror: Automatic Video to 3D Caricature Translation
- 3DCaricShop: A Dataset and A Baseline Method for Single-view 3D Caricature Face Reconstruction
- Attribute-Driven Spontaneous Motion in Unpaired Image Translation
- MangaGAN: Unpaired Photo-to-Manga Translation Based on The Methodology of Manga Drawing
- FA-GANs: Facial Attractiveness Enhancement with Generative Adversarial Networks on Frontal Faces
- Unsupervised Domain Attention Adaptation Network for Caricature Attribute Recognition
- CariMe: Unpaired Caricature Generation with Multiple Exaggerations
- Few-shot Knowledge Transfer for Fine-grained Cartoon Face Generation
- Unsupervised Contrastive Photo-to-Caricature Translation based on Auto-distortion